0.2

Adjusted Signal

0.2

Raw Pattern Score

100%

Time Factor

2026-10-30

Last Push

24.2K

Stars

Language

5.0M

Lines of Code

3.3K

Files

438

Pattern Hits

2026-10-31

Scan Date

0.03

HC Hit Rate

Risk Review · Payload v1

Use each vector to guide manual verification of the underlying findings.

Payload schema: 1.0.0 · Engine: 2.0.0 · Scoring: heuristic-density-v1 · Scan date: 2026-10-31.

Cognitive Ownership Risk 0.3/100 · LOW

Can maintainers explain intent, assumptions, and trade-offs? 292 finding(s).

Hallucinated Dependency & Supply Chain Risk 0.6/100 · LOW

Are imports declared and package/API names verified? 131 finding(s).

Agent Over-Engineering Ratio 0.1/100 · LOW

Are wrappers, guards, and repeated structures necessary? 106 finding(s).

PR Review Velocity Impact 1.0/100 · LOW

Which changes need focused human review? 438 finding(s).

What These Metrics Mean

Adjusted Signal: Legacy pattern score normalized per 1,000 scanned lines and multiplied by the repository-level time factor. It is not an estimate of AI authorship.
Raw Pattern Score: Sum of severity- and context-weighted pattern findings before applying the time factor.
Time Factor: A coarse multiplier based on repository last-push metadata. It does not identify when individual code was written.
Pattern Hits: Count of configured pattern findings. Review each location and surrounding code before acting.
HC Hit Rate: High- and critical-severity findings per scanned file. Severity is a configured detector weight, not proof of a vulnerability or authorship.
Lines / Files: Scan coverage denominators used by the legacy normalized score.

Pattern-Signal History

This chart tracks the legacy adjusted pattern score across scan runs. Changes can reflect source changes, scan coverage, or detector/scoring versions; they do not measure AI adoption or prove that a team refactored code. Compare runs only when their scoring versions are compatible. The dashed line tracks the raw pattern-hit count.

Severity Breakdown

Groups findings by configured severity weight. Severity is not a confidence estimate, security rating, or proof that a package is missing. Verify imports against the project's manifests and inspect each finding in its local design context.

CRITICAL 0HIGH 108MEDIUM 34LOW 296

Directory Score Breakdown

This horizontal bar chart shows where configured pattern scores accumulate across directories. Treat the bars as locations to inspect, not as evidence of authorship or code quality. Use the file-level findings to decide whether a directory needs a focused ownership, dependency, complexity, or review check.

Pattern Findings

The scanner identified 438 distinct pattern matches across 13 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a detector match to inspect in context. Expand any category row to inspect the individual file paths, line numbers, code snippets, and the lexical context (CODE, COMMENT, or STRING) in which each match was detected.

Reading the findings table: The Severity column indicates the configured finding weight (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — legacy comment-context matches receive a ×1.5 score weight. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier under the legacy scoring model.

Cross-Language Confusion105 hits · 539 pts

Severity	File	Line	Snippet	Context
HIGH	tests/test_dspark_window.py	111	if (pend_n && pend_start + pend_n == L) catch_up(&g, pend_start, pend_n);	CODE
HIGH	tests/test_dspark_window.py	157	REQUIRE(g.dspark_cache_token_start == 76 && g.dspark_cache_len == 127);	CODE
HIGH	tests/test_dspark_window.py	159	REQUIRE(g.dspark_cache_token_start == 76 && g.dspark_cache_len == 128);	CODE
HIGH	tests/test_dspark_window.py	184	if (start != 0 && p0 % cap) {	CODE
HIGH	tests/test_dspark_window.py	189	if (p1 < end && end - p1 < 128u) seed_tail(g, p0, chunk);	CODE
HIGH	tests/test_dspark_window.py	256	REQUIRE(g.dspark_cache_token_start == 0 && g.dspark_cache_len == 59);	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	127	if(s!=active_session \|\| f->frontier!=start \|\| s->checkpoint.len!=start \|\|	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	128	rows<1 \|\| rows>DS4_DSPARK_MAX_BLOCK_SIZE) mock_errors++;	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	131	if(s!=active_session \|\| !s->checkpoint_valid \|\| s->checkpoint.len!=s->graph.frontier) mock_errors++;	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	143	if(!tp_leader \|\| ctx!=active_session->engine->tp.ctx \|\| id!=42 \|\|	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	144	active_session->checkpoint.len!=tp_start \|\| active_session->graph.frontier!=tp_start) mock_errors++;	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	145	if(n==0 \|\| n>16) { mock_errors++; return 0; }	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	204	if(g!=&active_session->graph \|\| w!=&active_session->engine->weights \|\| !n \|\| rows) mock_errors++;	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	210	bytes!=row_bytes \|\| offset%row_bytes \|\| offset/row_bytes>=(uint64_t)verified_width) {	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	220	row!=(uint32_t)selected_row \|\| n!=DS4_N_VOCAB/2) { mock_errors++; return 0; }	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	338	if(tp_proposed && width>(already==0 ? 1 : 0)) {	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	347	if(expected==width && width) {	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	348	if(replayed) CHECK(tp_commit_kind==DS4_TP_VERIFY_ROLLBACK_REPLAY && tp_commit_count==expected);	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	349	else CHECK(tp_commit_kind==DS4_TP_VERIFY_COMMIT_FULL && tp_commit_count==0);	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	351	if(expected<width && expected && prefix_committed) {	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	352	CHECK(tp_commit_kind==DS4_TP_VERIFY_COMMIT_PREFIX && tp_commit_count==expected);	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	354	if(expected<width && expected && !prefix_committed) {	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	355	CHECK(tp_commit_kind==DS4_TP_VERIFY_ROLLBACK_REPLAY && tp_commit_count==expected);	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	357	if(width && !expected) CHECK(tp_commit_kind==DS4_TP_VERIFY_ROLLBACK_REPLAY && tp_commit_count==0);	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	358	CHECK(replay_half_reads==((replayed && engine.tp.vocab_split) ? 1 : 0));	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	425	int expected=(reject>=0 && reject<bounded) ? reject : bounded;	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	426	int width=(bounded==0 \|\| reject==0) ? 0 : bounded;	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	431	if(expected==width && width) CHECK(selected_row==expected-1);	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	432	if(eos<0 \|\| ignore) unaffected_failures+=failures-previous_failures;	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	494	if(width==1 && seed==0) worker_proposal_valid=false;	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	498	CHECK(head_local_reads==split && head_remote_reads==split);	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	507	int reads=(reject<0 \|\| (reject>0 && reject<=4)) ? split : 0;	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	508	CHECK(head_local_reads==reads && head_remote_reads==reads);	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	514	CHECK(head_local_reads==split && head_remote_reads==0);	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	521	CHECK(returned==-1 && tp_sends==1 && tp_width==0 && verified_width==0);	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	522	CHECK(session.checkpoint.len==start && session.graph.frontier==start && session.checkpoint_valid);	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	523	CHECK(!session.dspark_tp_proposal && !session.dspark_draft_valid && session.dspark_draft_len==0);	CODE
HIGH⚡	tests/test_dspark_eos_contract.py	524	CHECK(accepted[0]==-99 && mock_errors==0); cases++;	CODE
HIGH	tests/test_dspark_eos_contract.py	108	(void)e; (void)think; return token == 1 \|\| token == extra_stop;	CODE
HIGH	tests/test_dspark_eos_contract.py	114	if(t->len<0 \|\| t->len>=64) { mock_errors++; return; } t->data[t->len++]=token;	CODE
HIGH	tests/test_dspark_eos_contract.py	158	if(ctx!=active_session->engine->tp.ctx \|\| n!=DS4_N_VOCAB/2 \|\| !replayed) { mock_errors++; return 0; }	CODE
HIGH	tests/test_dspark_eos_contract.py	175	if(n==0 \|\| n>16 \|\| tokens->len!=(int)(start+n) \|\| g->frontier!=(int)start) mock_errors++;	CODE
HIGH	tests/test_dspark_eos_contract.py	231	if(n==0 \|\| n>4 \|\| n>=(uint32_t)verified_width) { mock_errors++; return false; }	CODE
HIGH	tests/test_dspark_eos_contract.py	245	if(g->frontier!=(int)pos \|\| token<0 \|\| token>=DS4_N_VOCAB) { mock_errors++; return false; }	CODE
HIGH	tests/test_dspark_eos_contract.py	279	session.logits[(draft[0]>=0 && draft[0]<DS4_N_VOCAB) ? draft[0] : 0]=100;	CODE
HIGH	tests/test_dspark_eos_contract.py	321	CHECK(!session.dspark_draft_valid && session.dspark_draft_len==0);	CODE
HIGH	tests/test_dspark_eos_contract.py	326	CHECK(accepted[cap]==-99 && accepted[19]==-99);	CODE
HIGH	tests/test_dspark_eos_contract.py	369	if(tp && n>seed) expected\|=DS4_TP_VERIFY_DSPARK_TP \| (seed ? DS4_TP_VERIFY_DSPARK_SEED : 0);	CODE
HIGH	tests/test_dspark_eos_contract.py	412	observed_negative=(returned==2 && session.checkpoint.len-start==3);	CODE
HIGH	tests/test_dspark_eos_contract.py	472	returned=invoke(8,8,false); check_frontier(returned,5,6,extended,8); CHECK(prefix_committed==4 && replayed==1);	CODE
HIGH	tests/test_dspark_eos_contract.py	536	CHECK(head_local_reads==1 && head_remote_reads==1 && replayed==1 && replay_half_reads==1);	CODE
HIGH	tests/test_dspark_confidence_input.py	103	if (!t \|\| off > t->bytes \|\| n > t->bytes - off) return 0;	CODE
HIGH	tests/test_dspark_confidence_input.py	130	if (!t \|\| t->id != T_W1 \|\| row >= DS4_N_VOCAB) return false;	CODE
HIGH	tests/test_dspark_confidence_input.py	221	REQUIRE(g_reads_cur == 1 && g_reads_norm == 0);	CODE
HIGH	tests/test_dspark_confidence_input.py	233	REQUIRE(plen == BLOCK && clen == BLOCK && cprefix == BLOCK);	CODE
HIGH⚡	tests/test_dspark_rewind.py	132	static bool ds4_session_tp_leader(ds4_session *s) { return s->engine->tp.active && s->leader; }	CODE
HIGH⚡	tests/test_dspark_rewind.py	140	MOCK(f==&s->dspark_rewind_frontier && f->epoch==s->epoch);	CODE
HIGH⚡	tests/test_dspark_rewind.py	142	if(wire.in_recent && ds4_session_tp_leader(s)) MOCK(wire.send_calls==1 && wire.wait_calls==0);	CODE
HIGH⚡	tests/test_dspark_rewind.py	149	MOCK(n>0 && n<=DS4_SPEC_PREFIX_SLOTS && n<=s->dspark_rewind_prefixes);	CODE
HIGH⚡	tests/test_dspark_rewind.py	150	MOCK(wire.restored[s->rank]==1 && s->graph.target_pos==s->dspark_rewind_start);	CODE
45 more matches not shown…

Over-Commented Block112 hits · 112 pts

Severity	File	Line	Snippet	Context
LOW	rax_malloc.h	41	#define rax_realloc realloc	COMMENT
LOW	ds4_ssd.c	1	#include "ds4_ssd.h"	COMMENT
LOW	ds4_agent.c	1	#include "ds4.h"	COMMENT
LOW	ds4_agent.c	21	#include <regex.h>	COMMENT
LOW	ds4_kvstore.h	1	#ifndef DS4_KVSTORE_H	COMMENT
LOW	ds4_distributed.c	21	#include <math.h>	COMMENT
LOW	ds4_distributed.c	41	* Protocol Constants And Wire Records	COMMENT
LOW	ds4_tp.c	1	/* Tensor-parallel transport and lockstep protocol. See ds4_tp.h.	COMMENT
LOW	ds4_tp.c	21	#include <stdlib.h>	COMMENT
LOW	ds4_tp.c	61	#define DS4_TP_MBOX_ACK_EVERY 16u	COMMENT
LOW	ds4_rocm.h	1	#pragma once	COMMENT
LOW	ds4_rocm.h	21	#define cudaGetErrorString hipGetErrorString	COMMENT
LOW	ds4_rocm.h	41	#define cudaMemcpyAsync hipMemcpyAsync	COMMENT
LOW	ds4_rocm.h	61	#define cudaStreamDestroy hipStreamDestroy	COMMENT
LOW	ds4_rocm.h	81	#define CUBLAS_COMPUTE_32F HIPBLAS_COMPUTE_32F	COMMENT
LOW	ds4_server.c	1	#include "ds4.h"	COMMENT
LOW	ds4_server.c	21	#include <errno.h>	COMMENT
LOW	ds4_server.c	5841		COMMENT
LOW	ds4_image.c	1	#include "ds4_image.h"	COMMENT
LOW	ds4_cli.c	1	#include "ds4.h"	COMMENT
LOW	linenoise.c	101	* Sequence: ESC [ 2 J	COMMENT
LOW	ds4_engram.c	1	#define _DARWIN_C_SOURCE	COMMENT
LOW	ds4_kvstore.c	21	#include <sys/types.h>	COMMENT
LOW	ds4_web.c	1	#include "ds4_web.h"	COMMENT
LOW	ds4_eval.c	21	#include <ctype.h>	COMMENT
LOW	cuda/mmq/ggml-common.h	21	// std-c++ do not allow it.	COMMENT
LOW	cuda/mmq/ggml-common.h	61	#elif defined(GGML_COMMON_DECL_SYCL)	COMMENT
LOW	cuda/mmq/ggml-common.h	81	#define static_assert(cond, msg) struct global_scope_noop_trick	COMMENT
LOW	cuda/mmq/ggml-common.h	101	#define QR4_0 2	COMMENT
LOW	cuda/mmq/ggml-common.h	121	#define QI8_1 (QK8_1 / (4 * QR8_1))	COMMENT
LOW	cuda/mmq/ggml-common.h	141		COMMENT
LOW	cuda/mmq/ggml-common.h	161	#define QR4_NL 2	COMMENT
LOW	cuda/mmq/ggml-common.h	461		COMMENT
LOW	cuda/mmq/ggml-common.h	481	#include <cstdint>	COMMENT
LOW	cuda/mmq/ds4_ggml_stubs.h	1	// SPDX-License-Identifier: MIT	COMMENT
LOW	cuda/mmq/ds4_ggml_stubs.h	21	// GGML_CUDA_MAX_STREAMS / GGML_LOG_DEBUG	COMMENT
LOW	cuda/mmq/ds4_ggml_stubs.h	41	#include <cstdint>	COMMENT
LOW	cuda/mmq/ds4_ggml_stubs.h	61	abort(); \	COMMENT
LOW	cuda/mmq/ds4_ggml_stubs.h	81	#define GGML_UNUSED_VARS_11(_1,_2,_3,_4,_5,_6,_7,_8,_9,_10,_11) GGML_UNUSED_VARS_10(_1,_2,_3,_4,_5,_6,_7,_8,_	COMMENT
LOW	cuda/mmq/ds4_ggml_stubs.h	101	#define GGML_MAX_SRC 10	COMMENT
LOW	cuda/mmq/ds4_ggml_stubs.h	121	#undef GGML_CUDA_USE_GRAPHS	COMMENT
LOW	cuda/mmq/ds4_ggml_stubs.h	181	// ----------------------------------------------------------------------------	COMMENT
LOW	cuda/mmq/ggml-cuda.h	1	// SPDX-License-Identifier: MIT	COMMENT
LOW	cuda/mmq/ggml-impl.h	1	// SPDX-License-Identifier: MIT	COMMENT
LOW	cuda/mmq/ds4_mmq.h	1	// SPDX-License-Identifier: MIT	COMMENT
LOW	cuda/mmq/ds4_mmq.h	21	#endif	COMMENT
LOW	cuda/mmq/ds4_mmq.h	41	// Wraps ggml_cuda_should_use_mmq. type_x uses ds4 quant codes which match	COMMENT
LOW	cuda/mmq/ds4_mmq.h	181	// Layouts:	COMMENT
LOW	cuda/mmq/ds4_mmq.h	421	cudaStream_t stream);	COMMENT
LOW	cuda/mmq/ds4_mmq.h	501	cudaStream_t stream);	COMMENT
LOW	cuda/mmq/ds4_mmq.h	521		COMMENT
LOW	cuda/mmq/ds4_mmq.h	661	// out[token, row] = sum_slot W[ids[token, slot], row, :] @ X[token, slot, :]	COMMENT
LOW	cuda/mmq/ds4_mmq.h	721	//	COMMENT
LOW	cuda/mmq/ds4_mmq.h	781	//	COMMENT
LOW	cuda/mmq/ggml.h	1	// SPDX-License-Identifier: MIT	COMMENT
LOW	cuda/mmq/vendors/hip.h	1	#pragma once	COMMENT
LOW	cuda/mmq/vendors/hip.h	21	#define CUDA_R_16BF HIPBLAS_R_16B	COMMENT
LOW	cuda/mmq/vendors/hip.h	41	#define cublasGemmEx hipblasGemmEx	COMMENT
LOW	cuda/mmq/vendors/hip.h	61	#define cudaErrorMemoryAllocation hipErrorOutOfMemory	COMMENT
LOW	cuda/mmq/vendors/hip.h	81	#define cudaHostUnregister hipHostUnregister	COMMENT
52 more matches not shown…

Hyper-Verbose Identifiers91 hits · 102 pts

Severity	File	Line	Snippet	Context
LOW	tests/test_model_download.py	143	def test_quants_vision_and_existing(self):	STRING
LOW	tests/test_model_download.py	157	def test_truncated_and_corrupt_artifacts(self):	STRING
LOW⚡	tests/test_model_download.py	170	def test_qwen_is_one_verified_file(self):	STRING
LOW⚡	tests/test_model_download.py	180	def test_failure_does_not_replace_link(self):	STRING
LOW⚡	tests/test_model_download.py	187	def test_partial_requires_explicit_cleanup(self):	STRING
LOW	tests/test_model_download.py	245	def test_help_and_invalid_target(self):	STRING
LOW	tests/test_jpeg_decode.py	58	def test_sampling_progression_and_edges(self):	CODE
LOW	tests/test_deepseek41_vision_conversion.py	40	def test_reject_missing_extra_and_wrong_tensors(self):	CODE
LOW	tests/test_serve_concurrency_bench.py	28	def test_counts_tokens_not_chunks(self):	CODE
LOW	tests/test_serve_concurrency_bench.py	34	def test_rejects_incomplete_or_failed_streams(self):	CODE
LOW⚡	tests/test_qwen4_vision_ref.py	10	def test_grid_mismatch_with_identical_shape_fails(self):	CODE
LOW⚡	tests/test_qwen4_vision_ref.py	14	def test_nonfinite_reference_or_actual_fails(self):	CODE
LOW⚡	tests/test_qwen4_vision_ref.py	22	def test_one_bad_token_fails_despite_high_mean(self):	CODE
LOW⚡	tests/test_qwen4_vision_ref.py	31	def test_scaled_vectors_pass_cosine_but_report_absolute_error(self):	CODE
LOW	tests/test_deepseek41_conversion.py	192	def test_invalid_scales_and_float(self):	CODE
LOW	tests/test_collect_official.py	19	def test_temperature_and_record(self):	CODE
LOW	tests/test_collect_official.py	45	def test_system_message_and_resume(self):	CODE
LOW⚡	tests/test_glm_reference_render.py	19	def test_preserves_reasoning_and_whitespace(self):	CODE
LOW⚡	tests/test_glm_reference_render.py	27	def test_missing_reasoning_rejected(self):	CODE
LOW⚡	tests/test_glm_reference_render.py	31	def test_other_model_rejected(self):	CODE
LOW⚡	tests/test_glm_reference_render.py	37	def test_unsupported_effort_rejected(self):	CODE
LOW	tests/test_deepseek41_dspark_conversion.py	43	def test_draft_only_source_shards(self):	CODE
LOW	tests/test_deepseek41_dspark_conversion.py	132	def test_atomic_resume_and_complete_payload_audit(self):	CODE
LOW	gguf-tools/qwen4_pack.py	313	def decode_safetensors_header(raw_length: bytes, raw_header: bytes,	CODE
LOW	gguf-tools/qwen4_pack.py	402	def read_hub_safetensors_header(repo_id: str, revision: str, filename: str,	CODE
LOW	gguf-tools/qwen4_pack.py	1121	def default_quantizer_library() -> Path:	CODE
LOW	gguf-tools/qwen4_pack.py	2004	def verify_shard_payload_auth(source_shard: str, auth: dict,	CODE
LOW	gguf-tools/qwen4_pack.py	2219	def validate_artifact_tensor_counts(plan: dict[str, list[Action]],	CODE
LOW	gguf-tools/glm53_quantize.py	259	def validate_tokenizer_template(hf_dir, template_tokens, vocab_size):	CODE
LOW	gguf-tools/tests/test_glm53_quantize.py	90	def test_conversion_signature_includes_imatrix_contents(self):	CODE
LOW	gguf-tools/tests/test_glm53_quantize.py	102	def test_imatrix_unobserved_expert_uses_fallback(self):	CODE
LOW	gguf-tools/tests/test_glm53_quantize.py	121	def test_imatrix_accepts_full_glm_expert_count(self):	CODE
LOW	gguf-tools/tests/test_glm53_quantize.py	175	def test_full_glm_metadata_keys_are_unique(self):	CODE
LOW	gguf-tools/tests/test_glm53_quantize.py	257	def test_combined_kv_is_split_per_head_and_k_is_transposed(self):	CODE
LOW	gguf-tools/tests/test_glm53_quantize.py	269	def test_full_glm_combined_kv_dimensions(self):	CODE
LOW	gguf-tools/tests/test_qwen4_pack.py	288	def test_manifest_records_qtypes_physical_shapes_padding_and_checksums(self):	CODE
LOW	gguf-tools/tests/test_qwen4_pack.py	373	def test_legacy_artifacts_are_refused_even_for_a_fresh_conversion(self):	CODE
LOW	gguf-tools/tests/test_qwen4_pack.py	381	def test_streamed_conversion_refuses_finalized_output_directory(self):	CODE
LOW	gguf-tools/tests/test_qwen4_pack.py	412	def test_conversion_space_admission_counts_artifacts_stage_and_reserve(self):	CODE
LOW	gguf-tools/tests/test_qwen4_pack.py	431	def test_random_gguf_writer_is_resumable_and_offset_stable(self):	CODE
LOW	gguf-tools/tests/test_qwen4_pack.py	495	def test_completed_tensor_payload_corruption_is_not_resigned(self):	CODE
LOW	gguf-tools/tests/test_qwen4_pack.py	556	def test_ple_gguf_writer_resumes_and_preserves_rows_and_aux_tensors(self):	CODE
LOW	gguf-tools/tests/test_qwen4_pack.py	615	def test_completed_ple_row_and_aux_corruption_is_not_resigned(self):	CODE
LOW	gguf-tools/tests/test_qwen4_pack.py	803	def test_optional_mtp_uses_the_same_qtype_rules(self):	CODE
LOW	gguf-tools/tests/test_qwen4_pack.py	948	def test_qwen_gguf_imatrix_validates_normalizes_and_falls_back(self):	CODE
LOW	gguf-tools/tests/test_qwen4_pack.py	978	def test_qwen_gguf_imatrix_rejects_missing_or_negative_entries(self):	CODE
LOW	gguf-tools/tests/test_qwen4_pack.py	1000	def test_block_quantization_rejects_invalid_width_and_nonfinite_input(self):	CODE
LOW	gguf-tools/tests/test_qwen4_pack.py	1008	def test_single_base_contains_every_layer_and_preserves_locality(self):	CODE
LOW	gguf-tools/tests/test_qwen4_pack.py	1032	def test_sidecars_are_classified_before_single_base(self):	CODE
LOW	gguf-tools/tests/test_qwen4_pack.py	1051	def test_official_geometry_is_required(self):	CODE
LOW⚡	gguf-tools/tests/test_qwen4_native_ngrams.py	82	def test_reject_wrong_source_precision(self):	CODE
LOW⚡	gguf-tools/tests/test_qwen4_native_ngrams.py	89	def test_upstream_geometry_metadata(self):	CODE
LOW⚡	gguf-tools/tests/test_qwen4_native_ngrams.py	99	def test_source_padding_is_retained(self):	CODE
LOW⚡	gguf-tools/tests/test_qwen4_native_ngrams.py	118	def test_reuse_verified_native_gguf(self):	CODE
LOW⚡	gguf-tools/tests/test_qwen4_native_ngrams.py	126	def test_reject_wrong_hash_geometry(self):	CODE
LOW⚡	gguf-tools/tests/test_qwen4_native_ngrams.py	131	def test_reject_missing_shard(self):	CODE
LOW⚡	gguf-tools/tests/test_qwen4_native_ngrams.py	136	def test_reject_wrong_row_count(self):	CODE
LOW⚡	gguf-tools/tests/test_qwen4_native_ngrams.py	141	def test_reject_truncated_source(self):	CODE
LOW⚡	…/quality-testing/tests/test_compare_frontier_logits.py	59	def test_complete_identity_and_full_vector_hashes(self):	CODE
LOW⚡	…/quality-testing/tests/test_compare_frontier_logits.py	69	def test_non_argmax_drift_cannot_hide_behind_same_argmax(self):	CODE
31 more matches not shown…

Deep Nesting58 hits · 57 pts

Severity	File	Line	Context
LOW	tests/deepseek41_dspark_reference.py	97	CODE
LOW	tests/generate_long_context_story_prompt.py	134	CODE
LOW	tests/test_server_speculative_stop.py	64	CODE
LOW	tests/test_jpeg_decode.py	58	CODE
LOW	tests/test_dspark_window.py	35	CODE
LOW	tests/ds4_agent_terminal_test.py	58	CODE
LOW	tests/test_server_batching.py	170	CODE
LOW	tests/test_dspark_eos_contract.py	23	CODE
LOW	tests/test_web_recovery.py	13	CODE
LOW	tests/test_web_recovery.py	27	CODE
LOW	tests/test_qwen4_checkpoint_replay.py	41	CODE
LOW	tests/test_collect_official.py	45	CODE
LOW	tests/test_deepseek41_rope.py	15	CODE
LOW	tests/test_agent_vision.py	42	CODE
LOW	tests/test_deepseek41_http.py	14	CODE
LOW	tests/qwen4_vision_ref.py	31	CODE
LOW	tests/qwen4_vision_ref.py	86	CODE
LOW	tests/test_dspark_confidence_input.py	37	CODE
LOW	tests/test_dspark_rewind.py	25	CODE
LOW	tests/test-vectors/fetch_official_vectors.py	190	CODE
LOW	tests/test-vectors/fetch_openrouter_glm_vectors.py	276	CODE
LOW	gguf-tools/deepseek4_vision.py	294	CODE
LOW	gguf-tools/deepseek41_validate_gguf.py	22	CODE
LOW	gguf-tools/deepseek41_validate_gguf.py	54	CODE
LOW	gguf-tools/merge_deepseek41_shards.py	49	CODE
LOW	gguf-tools/merge_deepseek41_shards.py	176	CODE
LOW	gguf-tools/deepseek41_metadata.py	52	CODE
LOW	gguf-tools/deepseek41_dspark.py	196	CODE
LOW	gguf-tools/deepseek41_dspark.py	251	CODE
LOW	gguf-tools/deepseek41_quantize.py	218	CODE
LOW	gguf-tools/deepseek41_quantize.py	161	CODE
LOW	gguf-tools/qwen4_native_ngrams.py	30	CODE
LOW	gguf-tools/qwen4_native_ngrams.py	145	CODE
LOW	gguf-tools/gen_qwen4_unicode.py	20	CODE
LOW	gguf-tools/qwen4_pack.py	295	CODE
LOW	gguf-tools/qwen4_pack.py	360	CODE
LOW	gguf-tools/qwen4_pack.py	1256	CODE
LOW	gguf-tools/qwen4_pack.py	1671	CODE
LOW	gguf-tools/qwen4_pack.py	2402	CODE
LOW	gguf-tools/qwen4_pack.py	2779	CODE
LOW	gguf-tools/qwen4_pack.py	2878	CODE
LOW	gguf-tools/qwen4_pack.py	427	CODE
LOW	gguf-tools/glm53_vision.py	86	CODE
LOW	gguf-tools/glm53_vision.py	249	CODE
LOW	gguf-tools/glm53_quantize.py	190	CODE
LOW	gguf-tools/glm53_quantize.py	561	CODE
LOW	gguf-tools/glm53_quantize.py	1015	CODE
LOW	gguf-tools/glm53_quantize.py	784	CODE
LOW	gguf-tools/glm53_quantize.py	802	CODE
LOW	gguf-tools/qwen4_iq2.py	93	CODE
LOW	gguf-tools/quality-testing/compare_scores.py	18	CODE
LOW	gguf-tools/quality-testing/collect_official.py	225	CODE
LOW	gguf-tools/quality-testing/compare_frontier_logits.py	141	CODE
LOW	…quality-testing/generate_deepseek4_vision_reference.py	62	CODE
LOW	gguf-tools/imatrix/dataset/build_ds4_imatrix_dataset.py	157	CODE
LOW	gguf-tools/imatrix/dataset/build_ds4_imatrix_dataset.py	872	CODE
LOW	speed-bench/serve_concurrency_bench.py	68	CODE
LOW	dir-steering/tools/build_direction.py	117	CODE

AI Slop Vocabulary26 hits · 55 pts

Severity	File	Line	Snippet	Context
MEDIUM	…ek-v4.1-flash-20260910-general/responses/case_066.json	11	"content": "# Testing a Command-Line Parser with Edge Cases\n\nA robust CLI parser needs to handle malformed, am	CODE
MEDIUM	…eek-v4.1-flash-20260919-router/responses/case_019.json	11	"content": "# Handling Clients That Send Valid Headers But Never Finish the Request Body\n\nThis is a classic sl	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_076.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "Testare	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_001.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "Here is	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_000.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "Here is	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_061.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "Sorted	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_046.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "In a Mi	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_089.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "A hash	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_066.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "# Testi	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_066.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "# Testi	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_026.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "Here is	CODE
MEDIUM⚡	…/data/qwen38-flash-alibaba-100/responses/case_029.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "Compari	CODE
MEDIUM⚡	…/data/qwen38-flash-alibaba-100/responses/case_029.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "Compari	CODE
MEDIUM⚡	…/data/qwen38-flash-alibaba-100/responses/case_029.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "Compari	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_009.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "Here ar	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_048.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "Validat	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_090.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "In the	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_023.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "**Backp	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_003.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "To deri	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_003.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "To deri	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_002.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "Compari	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_080.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "A good	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_059.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "Valutar	CODE
MEDIUM	…/data/qwen38-flash-alibaba-100/responses/case_034.json	1	{"model": "qwen/qwen3.8-flash", "provider": "Alibaba", "choices": [{"message": {"role": "assistant", "content": "Salvare	CODE
MEDIUM	…3-flash-openrouter-zai-fp8-100/responses/case_029.json	17	"content": "# Comparing Two Model Quantizations Without Relying on a Single Answer\n\nEvaluating quantization qu	CODE
MEDIUM	…3-flash-openrouter-zai-fp8-100/responses/case_024.json	17	"content": "# Safe Format for Storing Model Checkpoint Headers\n\nA checkpoint header is the metadata at the sta	CODE

Unused Imports25 hits · 25 pts

Severity	File	Line	Context
LOW	metal/generate_mxfp4_half_lut.py	8	CODE
LOW	tests/generate_long_context_story_prompt.py	9	CODE
LOW	tests/test-vectors/fetch_official_vectors.py	9	CODE
LOW	tests/test-vectors/fetch_openrouter_glm_vectors.py	9	CODE
LOW	gguf-tools/glm53_full_quantize.py	4	CODE
LOW	gguf-tools/deepseek4_vision.py	4	CODE
LOW	gguf-tools/deepseek4_vision.py	16	CODE
LOW	gguf-tools/deepseek4_vision.py	16	CODE
LOW	gguf-tools/glm53_validate_gguf.py	4	CODE
LOW	gguf-tools/glm53_validate_gguf.py	10	CODE
LOW	gguf-tools/qwen4_pack.py	18	CODE
LOW	gguf-tools/glm53_vision.py	4	CODE
LOW	gguf-tools/glm53_vision.py	15	CODE
LOW	gguf-tools/qwen4_exp_convert.py	16	CODE
LOW	gguf-tools/qwen4_pack_to_qwen4exp.py	33	CODE
LOW	gguf-tools/qwen4_pack_to_qwen4exp.py	38	CODE
LOW	gguf-tools/glm53_quantize.py	4	CODE
LOW	gguf-tools/mixed/splice_mixed_expert_layers_gguf.py	10	CODE
LOW	gguf-tools/quality-testing/validate_scores.py	4	CODE
LOW	gguf-tools/quality-testing/compare_scores.py	4	CODE
LOW	gguf-tools/quality-testing/collect_official.py	4	CODE
LOW	gguf-tools/quality-testing/compare_frontier_logits.py	4	CODE
LOW	…quality-testing/generate_deepseek4_vision_reference.py	4	CODE
LOW	gguf-tools/imatrix/dataset/build_ds4_imatrix_dataset.py	18	CODE
LOW	dir-steering/tools/build_direction.py	13	CODE

Decorative Section Separators6 hits · 16 pts

Severity	File	Line	Snippet	Context
MEDIUM	gguf-tools/qwen4_pack_to_qwen4exp.py	149	# --------------------------------------------------------------------------	COMMENT
MEDIUM	gguf-tools/qwen4_pack_to_qwen4exp.py	461	# --------------------------------------------------------------------------	COMMENT
MEDIUM	gguf-tools/qwen4_pack_to_qwen4exp.py	592	# --------------------------------------------------------------------------	COMMENT
MEDIUM	gguf-tools/qwen4_pack_to_qwen4exp.py	687	# --------------------------------------------------------------------------	COMMENT
MEDIUM	…3-flash-openrouter-zai-fp8-100/responses/case_034.json	17	"content": "# Formato sicuro per l'header di un checkpoint modello\n\n## Struttura consigliata\n\nUn approccio r	CODE
MEDIUM	…ols/quality-testing/data/flash/responses/case_001.json	11	"content": "Here's a concise design for a TCP echo server that handles slow clients:\n\n## Core Architecture\n\n	CODE

Empty Error Handlers3 hits · 15 pts

Severity	File	Line	Snippet	Context
HIGH	gguf-tools/qwen4_pack.py	2770	except OSError:	CODE
HIGH	gguf-tools/qwen4_pack.py	2874	except OSError:	CODE
HIGH	gguf-tools/qwen4_pack.py	2965	except OSError:	CODE

Excessive Try-Catch Wrapping6 hits · 6 pts

Severity	File	Line	Snippet	Context
LOW	tests/test_server_speculative_stop.py	161	except Exception as exc:	CODE
LOW	tests/test_server_speculative_stop.py	239	except Exception as exc:	CODE
LOW	tests/test-vectors/fetch_openrouter_glm_vectors.py	185	except Exception as e: # noqa: BLE001 - command-line retry wrapper.	CODE
LOW	gguf-tools/mixed/splice_mixed_expert_layers_gguf.py	402	except Exception as exc:	CODE
LOW	gguf-tools/quality-testing/collect_official.py	216	except Exception as e: # noqa: BLE001 - command-line retry wrapper.	CODE
LOW	speed-bench/serve_concurrency_bench.py	367	except Exception as exc: # noqa: BLE001 - any failure here is fatal and worth showing	CODE

Modern AI Meta-Vocabulary2 hits · 4 pts

Severity	File	Line	Snippet	Context
MEDIUM	…ek-v4.1-flash-20260910-general/responses/case_004.json	11	"content": "# Why KV Cache Checkpointing Helps Agentic Sessions\n\n## The core problem: long context, repeated p	CODE
MEDIUM	…sion-exp-openrouter-novita-100/responses/case_001.json	17	"content": "Here's a concise design for a TCP echo server that handles slow clients:\n\n## Architecture Overview	CODE

Structural Annotation Overuse2 hits · 2 pts

Severity	File	Line	Snippet	Context
LOW	…ek-v4.1-flash-20260910-general/responses/case_005.json	11	"content": "# Binary Heap Push (Insert) Operation\n\n## Pseudocode\n\n```\nfunction push(heap, value):\n // S	CODE
LOW	…v4.1-flash-20260910-general/continuations/case_005.txt	7	// Step 1: Add new element at the	COMMENT

Verbosity Indicators1 hit · 1 pts

Severity	File	Line	Snippet	Context
LOW	…ek-v4.1-flash-20260910-general/responses/case_005.json	11	"content": "# Binary Heap Push (Insert) Operation\n\n## Pseudocode\n\n```\nfunction push(heap, value):\n // S	CODE

Fake / Example Data1 hit · 1 pts

Severity	File	Line	Snippet	Context
LOW	…lash-openrouter-zai-fp8-100/continuations/case_066.txt	15	- Values containing spaces: `--name "John Doe"`	CODE

antirez/ds4

Open-Source Code Quality, Security and Technical Debt Review